SA602A oscillator load capacitance
 

I'm trying to figure out the load capacitance of the oscillator pins 6
and 7 on the SA602A. I've got a 14.85MHz fundamental crystal with a
specified load capacitance of 13pF running about 1300Hz slow (with
27+22p). The datasheet doesn't say much about the oscillator pins.

The best hint I've found is about the SA605:

From AN1994 about the SA605:

Because the Colpitts configuration is for parallel resonance
mode, it is important to know, when ordering crystals, that
the load capacitance of the SA605 is 10pF.

That would suggest I'm way, WAY off, and my external load capacitors
should only amount to 3p!

The SA612A datasheet has an example where it specifies the crystal load
capacitance, but then it doesn't specify C1/C2 values. Then it has
another example with specific values for C1/C2 (which are 4.5p in series)
but does not mention the load capacitance of the 3OT crystal. One at
that freq made by CTS has a load capacitance of 13p, which would suggest
that the oscillator is about 8.5p. The 602A has a similar example with
another unspecified crystal...

--
Ben Jackson AD7GD

http://www.ben.com/

SA602A oscillator load capacitance
 

Ben,

When a crystal manufacturer specifies a load capacitance for a crystal, it
is usually a good indication that the crystal is meant to operate in the
'parallel resonant' mode. This is a very misleading term however, as the
quartz blank only has one pure resonance, and that is a series resonance.
However, the equivalent LCR circuit of this series resonant circuit will
look like a capacitive or inductive reactance at frequencies removed from
the series resonant frequency. At some frequency, the crystal will look
inductive enough to resonate an external capacitance of a value specified as
the 'load capacitance".

Your crystal will will form a parallel resonant circuit at the marked
frequency with an external capacitance of 13 pF. This is fairly low, as most
crystals are specified for 20-32 pf loads. If your circuit already has too
much capacitance, you can lower the net capacitance seen by the crystal by
either using a small coupling cap whose value in series with the circuit
capacitance equals the load capacitance, or you can put an inductor across
the crystal whose value in parallel with the circuit capacitance will result
in a net capacitance value equal to the 'load capacitance'. In the former
case, the low value of coupling cap could cause an AC voltage drop that
might prevent the circuit from oscillating. In the latter case the added
inductance could cause a spurious oscillation at an unintended frequency.

All things considered in your case, I'd use the series coupling approach.

Joe
W3JDR


"Ben Jackson" wrote in message
...
I'm trying to figure out the load capacitance of the oscillator pins 6
and 7 on the SA602A. I've got a 14.85MHz fundamental crystal with a
specified load capacitance of 13pF running about 1300Hz slow (with
27+22p). The datasheet doesn't say much about the oscillator pins.

The best hint I've found is about the SA605:

From AN1994 about the SA605:

Because the Colpitts configuration is for parallel resonance
mode, it is important to know, when ordering crystals, that
the load capacitance of the SA605 is 10pF.

That would suggest I'm way, WAY off, and my external load capacitors
should only amount to 3p!

The SA612A datasheet has an example where it specifies the crystal load
capacitance, but then it doesn't specify C1/C2 values. Then it has
another example with specific values for C1/C2 (which are 4.5p in series)
but does not mention the load capacitance of the 3OT crystal. One at
that freq made by CTS has a load capacitance of 13p, which would suggest
that the oscillator is about 8.5p. The 602A has a similar example with
another unspecified crystal...

--
Ben Jackson AD7GD

http://www.ben.com/

SA602A oscillator load capacitance
 

On 2006-07-11, W3JDR wrote:
"Ben Jackson" wrote in message
...
I'm trying to figure out the load capacitance of the oscillator pins 6
and 7 on the SA602A.

...suggests...oscillator is about 8.5p.

All things considered in your case, I'd use the series coupling approach.

I like that idea, since the series capacitance reduces the effect of
unknowns on the oscillator side. Using my estimate above and given my
existing caps, I tried a 33p in series and got to 14.849995MHz, which
is a vast improvement.

Thanks!

--
Ben Jackson AD7GD

http://www.ben.com/